Asymmetry in FTS and FTR

Table 4: Identification strategy for Flight to Risk shocks

variables TFP Value of Risk Real rate Surplus ratio Variable of interest shocks

FTR Strategy 1

Productivity + . − + .

Flight to Risk 0 + − + .

Monetary policy 0 . + − .

Demand . . . . .

Residual . . . . .

Notes: The SVAR for identifying Flight to Risk shocks has 5 variables: TFP, Value of risk (which is Equity price minus Bond price), Real rate, Surplus ratio (which is One minus the ratio of Consumption of services and Non-durable to Total consumption), and a business cycle variable of interest for e.g. Investments , Output, Consumption, Hours, CPI etc. On impact of the respective shock, + means the variable is restricted to be positive and −means that the variable is restricted to be negative for the impact horizon. The impact horizon is shock period plus one more period. Symbol 0 signifies a zero restriction when the response variable is restricted to not respond to the shock contemporaneously, and Dots.signify that the impact response of that variable is left unrestricted.

5.3 Asymmetry in FTS and FTR

One of the peculiar and significant features of Flight to Safety (FTS) or a sudden increase in preference for safer investments is that the market volatility and uncer-tainty experienced in such episodes are not reciprocated during its complementary market phenomenon of Flight to Risk (FTR). FTR is when investors prefer risky investments to safer bets. Such differential behaviour is based on the human psyche and behavioural biases that are out of this study’s scope. However, for the purpose of our analysis, it is interesting to consider whether the impact of FTS is reversed for shocks of opposite magnitude. An FTR shock series is developed by choosing a complementary identification to the one imposed in the benchmark study. Impulse response results from this series help uncover FTR shocks’ impact and keep the analysis relevant and comparable to the benchmark FTS model results.

The FTR shocks are identified by changing the sign of the Price of risk series, i.e. by calculating it as the difference between the Price of S&P 500 and the

Figure 15: Asymmetry between Flight to Safety and Flight to Risk

Notes: The figure shows impulse responses to Flight to Safety (solid) and Flight to Risk (diamonds) shocks and their 68% confidence bands (shaded). FTS shocks are identified with restriction strategy 1 in the benchmark VAR model. The FTR shocks are identified by imposing restriction strategy discussed in the Results section Table4.

Data: 1983:Q1 to 2019:Q3.

5 . 3 A s y m m e t ry i n F T S a n d F T R 67 10-year Treasury bond. This could be called a Value of Risk series or negative Price of risk series. Therefore an increase in the Value of Risk or Equity minus Bond price series occurs when Equities get more expensive compared to bonds, and a positive shock to this series is studied as Flight to Risk. The identification strategy imposed in identifying FTR shocks is also adjusted to account for the complementary changes to benchmark strategy 1. A positive shock in the Value of Risk series for the impact horizon is restricted to a fall in Real rates and an increase in the Surplus ratio. Both these sign restrictions are opposite in signs to the sign restrictions imposed on these variables in the benchmark strategy 1 for identifying FTS shocks. Similar to FTS shocks the FTR shocks are also restricted to be orthogonal to TFP and monetary policy shocks. Table4 on page 65presents the identification restrictions imposed in identifying FTR shocks. The impulse responses to FTR shocks, as shown in Figure 15 on page 66, highlight the asymmetry in the response of Flight to Safety and Flight to Risk shocks.

These impulse responses to FTR shocks are smaller in magnitude and short-lived compared to the impulse responses to FTS shocks. The increase in Equity minus Bond price (or a decrease in Bond minus Equity price) from an FTR shock leads to very slow adjustments in investments, hours, output, non-residential investments, inflation, and TFP and real rates when compared with their response to FTS shocks. The lack of negative response in Real rates is because monetary policy response is exogenous to the model. An increase in inflation after Flight to Risk does not warrant policy cuts from the inflation-targeting central bank. Real rates become negative when inflation stabilizes. The residential investment shows a lack of response, which is not that puzzling when considering that residential purchases are long-term decisions. These can be put away easily when faced with a Flight to Safety phenomenon, but they are not immediately put on board after a Flight to Risk. Housing is a long-term investment, and in the case of Flight to Risk shocks, the immediate response should be felt in more volatile (risky) but liquid options.

The residential investment increases after 5 years of risk-taking. There is also a counter-intuitive fall in surplus ratio after an FTR shock. However, it can be explained by the slow response of output and consumption in comparison to the more significant increase in consumption (of Non-Durables and Services) habits.

The increase in habits is also responsible for the relative price of investment goods in terms of consumption good getting into the negative territory. The surplus ratio slowly returns to its pre-shock levels. There is also feedback from FTR shocks to TFP, which reacts positively 5-8 quarters after the shock. Interestingly the response is delayed as it was in response to FTS shocks, highlighting the expectations channel through which the capital flight can impact productivity in an economy.